The GX-7 is Gelid's latest heatsink and the first heatsink in their Gamer line. We have used the Tranquillo Rev2 and its predecessor the Tranquillo from Gelid's Silent line of coolers for years now and have been very happy with their performance, so we have high hopes for this new heatsink. The GX-7 is intended to be a higher powered version of the Tranquillo and not it's direct replacement.
The specifications for the GX-7 and the Tranquillo are very similar, so lets first compare the manufacturer specifications for both parts:
|Gelid Rev2 Tranquillo||Gelid GX-7|
|Air Flow (CFM):||58 max||75.6 / 128.6|
|Bearing:||Hydro Dynamic Bearing||Nanoflux Bearing|
|Cable Length (mm):||500||500|
|DC Voltage (V):||12||12|
|Fan Dimensions (mm)||120 (l) x 120 (w) x 25 (h)||120 (l) x 120 (w) x 25 (h)|
|Fan Speed (RPM):||750 - 1500||600 - 1800|
|Heat Sink Dimensions (mm):||74 (l) x 125 (w) x 153 (h)||130(l) x 65 (w) x 159 (h)|
|INCLUDED:||GC-2 Thermal Compound & Installation Kit for AMD & Intel sockets||GC-2 Thermal Compound & Installation Kit for AMD & Intel sockets|
|Life time MTTF at 40C (h):||50,000||100,000|
|Noise Level (dBA):||12 - 25.5||10 - 26.8|
|Static Pressure (mmAq):||1.6||2.66|
|Number of LED||-||4|
From the manufacturer specifications, the GX-7 should be a much better cooler than the Tranquillo Rev2. The included fan has over twice the CFM, yet is only 1.3 dBa louder. It is slightly taller, but thinner as well so it may fit better in some chassis that are not very deep. Lastly, the GX-7 uses nanoflux bearings, which utilize magnets to keep the majority of the internal moving parts in the fan from touching using magnetic fields. More information on this type of bearing can be found on Gelid's product page for the fan itself.
Manufacturer specifications are often misleading however; especially with fans. We have often seen specifications of two fans, where one has higher CFM rating, yet when we tested the fans ourselves it was actually lower. For that reason, we will be doing some informal comparisons between the two included fans to determine the relative performance for ourselves.
To visually compare the two heatsinks, we took a few pictures of both heatsinks side-by-side. The GX-7 is on the left, and the Tranquillo Rev2 is on the right:
The mounting for the Gelid GX-7 is identical to the Tranquillo, which is a three-step process. First, install the bolts through the motherboard and secure with four nuts. Second, attach the appropriate mounting bracket to the bottom of the heatsink. After that, slide the heatsink over the bolts and secure with a second set of nuts. The second set of nuts include a spring tension, so you do not have to worry as much about getting all the nuts tightened exactly the same.
This type of mount is very secure, although it requires that the fan is not mounted in order to fasten the second set of nuts. For this reason, it is suggested that you mount the heatsink before installing the motherboard into the chassis.
The fan itself is held in place with two wire clips. While not 100% secure, it works pretty well, although we recommend reinforcing the clips slightly if the system is going to be transported a long distance.
Performance Cooling and Noise
The blue LED's are not indirectly visible in a
well-lit chassis, but glow nicely when
in a darker environment.
At 7V, which is roughly the same RPM as what the fan will be running when a system is idling, the GX-7 fan is just a touch louder than the Tranquillo fan, although the airflow is nearly identical. Turning the voltage up to full speed at 12V, the GX-7 fan is noticeably louder than the Tranquillo fan. It does move a good deal more air at the same time though, so it is possible that this fan will not have to ramp all the way up to full speed in order to keep the heatsink cool. In a real-world scenario, the Gelid GX-7 will likely be noticably louder than the Tanquillo Rev2.
The manufacturer specifications claim that this fan can move twice the air as the Tranquillo fan, but that does not seem to be the case based on our informal testing. It moved a decent amount more air, but it was more in the range of 20-30% more.
To test the thermal performance of this cooler with the included fan, we recorded the temperature at both idle and load with the hardware listed below. We will be testing both the Gelid heatsinks, as well as testing a closed-loop liquid cooler (Coolit Eco II) as a comparison.
|Motherboard:||Asus Rampage III Formula|
|CPU:||Intel Core i7 980X|
|RAM:||3x Kingston HyperX DDR3-1600 4GB|
|GPU:||2x Nvidia GTX 480 (configured for SLI)|
|CPU Cooler:||Gelid Tranquillo Rev2 / Gelid GX-7 / Coolit Eco II|
|PSU:||Corsair HX 850W|
|Hard Drive:||Western Digital Raptor 150GB|
|Case:||Antec P183 V3 (case fans on low)|
To fully load the system, we will run a combination of Prime95 and Furmark until the temperatures stabilize (roughly 10 minutes) with CoreTemp and GPU-Z being used to log the temperatures. This will cause higher temperatures than any normal user is likely to see, but should give us a clear worse-case scenario for cooling.
|CPU Temperature||Gelid Tranquillo Rev2||Gelid GX-7||Coolit Eco II|
|Idle||31° C||29° C||36° C|
|Load||75° C||62° C||57° C|
At idle, the temperature for the GX-7 are slightly better than the Tranquillo Rev2, but not by a large margin. Both Gelid coolers however have a much lower idle temperature than the Coolit Eco II. Given our fan testing, we expected this cooler to be fairly quiet at idle, but when actually installed onto the Intel 980X CPU, the noise from the cooler was definitely noticeable.
Once we had the system under a full load, we saw an impressive 13° C improvement in temperatures over the Tranquillo Rev2, although the Coolit Eco II still had the advantage by 5° C. The fan was running at full speed at that point, so it was making a good amount of noise. In terms of noise, this cooler is really more on par with the Coolit ECO II, and not the Gelid Tranquillo.
Our testing so far has shown that the GX-7 provide much lower temperatures than the Tranquillo, but we do not yet know if that is because the heatsink is more efficient or if the more powerful fan is what is causing the difference. So for this set of testing, we are going to take the fan out of the equation.
There are two main reasons why you would replace the fan that comes with a cooler. Either you want more airflow, or you want quieter operation. Since Puget Systems has a strong focus on quiet computers, we almost always want to get quieter operation. One example of when we do this is our Serenity line of computers by replacing the stock Tranquillo Rev2 fan with a Scythe Slip Stream 1300RPM 120mm PWM Fan.
Since hot systems may have thermal issues running a quieter fan, we will have to change our testing system to match one of our Serenity qualified configurations. The Coolit Eco II is not intended for super-quiet operation, so we will be testing the GX-7 and the Tranquillo only this time.
|Motherboard:||Asus P8H67-M EVO REV 3.0|
|CPU:||Intel Core i5 2400 QUAD CORE 3.1GHz 95W|
|RAM:||2x Kingston DDR2-1333 4096MB|
|GPU:||Asus Geforce GT 430 1GB Silent|
|CPU Cooler:||Gelid Tranquillo Rev2 / Gelid GX-7|
|Hard Drive:||Intel 320 120GB SATA II 2.5inch SSD|
|Case:||Antec Mini P180 (Gunmetal)|
To fully load the system, we will again run a combination of Prime95 and Furmark until the temperatures stabilize (roughly 10 minutes) with CoreTemp and GPU-Z being used to log the temperatures. This will cause higher temperatures than any normal user is likely to see, but should give us a clear worse-case scenario for cooling.
|CPU Temperature||Gelid Tranquillo Rev2||Gelid GX-7|
|Idle||33° C||32° C|
|Load||57° C||57° C|
Test results just don't get any closer than this. We only saw a single degree of difference at idle, and that can easily be attributed to normal testing variances. These results show that when paired with a low-airflow fan, the Gelid GX-7 does not have any performance advantage over the Tranquillo.
Overall, the GX-7 performs very well. The included fan is rated for 100,000 hours, which is a staggering 11.5 years of continuous use. It is a bit loud compared to the Tranquillo Rev2, but Gelid is not touting it as a quiet solution so that is not a major drawback. However, if the additional noise level is acceptable, then why not use the Coolit ECO II? The Gelid GX-7 is roughly as loud as the Coolit Eco II, yet performs slightly below it..
The answer largely depends on how you can configure the cooling in your chassis. The Coolit Eco II only performs at optimal levels when it can be configured as an intake with another fan acting as an exhaust in the same area (usually a top fan). If the chassis you are using cannot accommodate that configuration, the Gelid GX-7 will easily out-perform the Coolit Eco II. While leaks are rare with closed-looped liquid cooling solutions like the Eco II, they can still happen, so the GX-7 also has the advantage in that regard.
From the viewpoint of Puget Systems, we don't have any problems with this cooler, but there is not currently a need for it in our product line. It certainly is a great looking cooler, but since the GX-7 has identical performance with a quiet 120mm fan, there is no performance benefit to switching to the more expensive GX-7 cooler for our quiet systems that currently use the Tranquillo Rev2. Likewise, all of our high-power systems use chassis that allow the Coolit Eco II to be configured as an intake, giving us better cooling for roughly the same noise level.